Arthritis is a late manifestation of Lyme disease. Allowed to progress untreated, Lyme arthritis can result in a severe, erosive arthritis. The majority of patients improve with antibiotic therapy, but a small percentage will continue to have persistent arthritis. Whether the cause of this post-treatment arthritis is due to persistence of the organism or autoimmune disease is an area of intense debate. The mechanisms by which B. burgdorferi can cause cartilage degradation have not been well studied. Unlike other bacteria that cause septic arthritis, B. burgdorferi does not secrete any enzymes capable of digesting extracellular matrix proteins in the joint. Matrix metalloproteinases (MMPs) are host enzymes that are capable of digesting multiple components of human cartilage. We have found evidence that MMPs are elevated in the synovial fluid of patients with Lyme arthritis. The pattern of MMP induction in patients with untreated and persistent disease is significantly different, suggesting that they may occur through different mechanisms. We have found that B. burgdorferi stimulates MMP induction from chondrocytes in a pattern similar to that found in patients with untreated Lyme arthritis, but dissimilar to that seen in patients after antibiotic therapy. The presence of B. burgdorferi results in degradation of cartilage explants in vitro. MMP inhibitors can block this degradation. Our main hypothesis is that cartilage erosions in patients with Lyme arthritis occurs through the induction of MMPs from joint tissue by contact with B. burgdorferi while arthritis in patients post-antibiotic therapy is due to activation of different MMPs through different pathways (e.g. autoimmune). In this proposal, we will take advantage of a large sample bank of patients with untreated and post-treatment Lyme arthritis to fully characterize MMP and cytokine expression in the joints. A full understanding of the MMP profiles in these patients will allow us to make comparisons between patients with untreated and post-treatment Lyme arthritis as well as to patients with other arthritides and attempt to correlate specific MMPs with disease manifestations. We will also correlate our findings with studies of MMPs in the mouse joint. Using the mouse model of Lyme arthritis, we will study the role of specific cell types in MMP production in the joint. Then, having identified MMPs potentially relevant in Lyme arthritis, we will examine the effects of specific MMP inhibitors or genetic knockouts of MMP genes on the course of murine arthritis. Finally, we will determine signaling pathways activated by B. burgdorferi that result in induction of pathogenic MMPs. Recent investigations have identified a group of receptors, designated Toll-like receptors (TLRs), as important pattern recognition receptors for bacterial products. Using TLR blocking antibodies, dominant negative cell lines and knockout mice, we will dissect the signaling pathways that lead to MMP induction. We believe that studies of MMPs in Lyme disease will lead to a better understanding of the pathogenesis of untreated and persistent Lyme arthritis. In addition, the presence of a clear initiator and excellent animal models of disease allow us to define the role of MMPs in Lyme arthritis in a manner not possible for other arthritides such as rheumatoid arthritis and may subsequently lead to the development of more specific therapies.